Therapeutic Efficacy and Safety of onasemnogene abeparvovec Gene Therapy for Spinal Muscular Atrophy Type 1: Updated Systematic Review with Meta-Analysis

doi:10.21203/rs.3.rs-8032586/v1

Objectives:

The objective of this study is to conduct a meta-analysis to evaluate the efficacy and safety of onasemnogene abeparvovec (OA).

Background:

Onasemnogene abeparvovec is a gene therapy approved from the U.S. Food and Drug Administration in May 2019. Unlike other therapies, it offers unique benefit of a single-time dose administration. The therapy has since been approved for the treatment of SMA1 patients in multiple countries. Given its increasing global use, an updated systematic analysis is warranted to comprehensively assess its safety and clinical efficacy.

Methods:

An electronic literature search was done across PubMed, Scopus and WOS databases until October 2025. We conducted a meta-analysis of single-arm studies investigating the effects of onasemnogene therapy on safety and efficacy outcomes. Safety outcomes include overall, serious and drug related adverse events while efficacy outcomes include overall and event-free survival and change from baseline in CHOP-INTEND score. Effect estimates were presented in random effect model as single proportions for dichotomous data and pooled mean change from baseline for continuous data with and 95% confidence intervals (CI) for both.

Results:

Overall, twenty-one studies were included with total of 565 SMA1 patients. The pooled percentage of overall survival was 98% [95% CI: 96:99]. Subgroup analysis based on previous treatment with other disease-modifying agents showed significant subgroup difference favoring those who were treated before (P-value = 0.0982). However, subgrouping according to dose (standard versus high) did not show significant subgroup difference (P-value = 0.2439). The pooled percentage of event-free survival was 78% [95% CI: 66:87]. Subgroup analysis based on previous treatment with other disease-modifying therapies did not demonstrate significant subgroup difference (P-value = 0.3313). However, subgrouping according to dose exhibited significant subgroup difference favoring high dose (P-value = 0.0005). Overall, serious, and drug-related adverse effects showed pooled proportion of 94% [95% CI: 75:100], 30% [95% CI: 9:57], and 63% [95% CI: 49:75], respectively. Thrombocytopenia was the most frequent adverse event. For the change from baseline in CHOP-Intend score, the pooled effect estimate was 15.77 [95% CI:12.07:19.47], and subgrouping according to previous treatment with disease-modifying agents showed significant subgroup difference (P-value = 0.0817).

Conclusion:

Onasemnogene improves chances of survival in SMA1 patients, especially if patients had previous treatment, significantly improves motor abilities, and is generally tolerable.

Neurology

Pediatrics

Spinal muscular atrophy type 1 (SMA1)

Onasemnogene abeparvovec

gene replacement therapy

Overall survival

Event-free survival

CHOP-INTEND

Updated systematic review

Meta-analysis

Spinal muscular atrophy (SMA) is a rare autosomal recessive progressive neuromuscular disorder that affects the alpha motor neurons in the brain stem and anterior horn of the spinal cord. Clinical severity varies according to genetic and other factors[1]. Spinal muscular atrophy type 1 (SMA1) is the most common and severe variant that usually manifests early on before 6 months of age. It is caused by biallelic loss-of-function mutation in survival motor neuron 1 gene (SMN1) located on chromosome 5; leading to compensatory expression of SMN2 gene that encodes a smaller fraction of truncated yet functional SMN protein. SMA1 patients usually have 2 copies of SMN2 gene. Affected infants usually cannot attain ability to sit independently with progressive severe muscular weakness, muscle atrophy, tongue fasciculations, bulbar weakness and respiratory failure leading to premature death[2]. Prevalence and incidence of SMA is estimated to be 1–2 per 100,000 people and 1 per 10,000 live births respectively with SMA1 accounting for approximately 60% of cases[3].

Onasemnogene abeparvovec is the first gene replacement therapy for SMA1 patients. It has been approved by US Food and Drug Administration (FDA) in May 2019 for the treatment of SMA1 patients who are younger than two years of age with up to 3 copies of SMN2 gene[4, 5]. It is an adeno-associated virus serotype 9 (AAV9) vector that carries a viable copy of self-complementary DNA of SMN1 gene and transmits it to the central nervous system due to the ability of the virus capsid to penetrate the blood brain barrier[6, 7]. The hybrid cytomegalovirus enhancer and chicken beta actin promoter accounts for the rapid, sustainable and ubiquitous expression of the gene. The drug is usually administered as one-time intravenous infusion[7]; However, intrathecal administration has been explored in some studies and is currently under investigation in ongoing trials (STEER phase III and Strength phase III[8–10]). Clinical efficacy and safety of the drug have been explored in multiple trials and real-world observational studies[11–15]. Additionally, other disease modifying agents (nusinersen and risdiplam) that alter the splicing of SMN2 gene to increase the production of fully functional form of SMN protein, have also been investigated and shown to improve motor function and survival of SMA1 patients[16–18].

Notably, the latest meta-analysis[19] that explored efficacy and safety of onasemnogene abeparvovec in SMA1 patients included only clinical trials published till November 2022[11–13]. However, recently, there has been a lot of real-world observational evidence describing the experience of onasemnogene abeparvovec in SMA1 patients, though, studies were limited by low statistical power due to limited sample size. Subsequently, we aimed to conduct this updated systematic review and meta-analysis to demonstrate a more robust real experience-based investigation of safety and effects of onasemnogene abeparvovec on survival and motor abilities of SMA1 patients with subgroup analysis according to dose of the drug and previous treatment with other disease-modifying therapies.

This systematic review and meta-analysis was conducted in concordance with the Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA) statement[20]. A protocol was pre-registered in the Prospero database (CRD420251164008).

Eligibility criteria:

The included studies should meet the following criteria:

Studies investigating Onasemnogene abeparvovec (Zolgensma) therapy whether they are Randomized Controlled Trials (RCTs), non-RCTs or observational studies including cohort studies, case series, or other forms of observational research.
Studies including pediatric patients diagnosed with SMA type 1 based on bi-allelic mutations in the SMN1 gene SMN2 copy number., less than 2 years of age at the time of SMA diagnosis and not requiring permanent ventilation when starting Onasemnogene therapy.
Studies including patients irrespective of their prior treatment (previously treated or treatment-naïve) or symptomatic status at the time of treatment (pre-symptomatic or post-symptomatic).
Studies must report data on at least one of the primary outcomes: Drug adverse events, overall survival, event-free survival and motor function improvement (as assessed by HFMS, RHS, or CHOP INTEND).

Non-English literature, reviews, case reports, book chapters, conference abstracts, editorials, opinion papers, qualitative studies, and animal studies were excluded.

Search strategy:

An electronic literature search was conducted using PubMed, Scopus and WOS databases. Databases were searched from time of inception until October 2025 using the following terms: "spinal muscular atrophy type 1" and "onasemnogene abeparvovec". Detailed search strategy is demonstrated in the supplementary file, Table S1. Duplicated studies were removed using Endnote software[21].

Study selection and data extraction:

Two independent authors conducted title and abstract screening followed by full text screening of the retrieved studies using Rayaan online software[22]. A third author was involved in resolving conflicts in each screening stage. Six authors, divided into two groups, conducted data extraction independently and then any conflicts were resolved by discussion or by independent third party. All data conversions, including transformation of medians (Interquartile range/range) to means (standard deviation) and other reported formats (e.g., combining individual patient data), were performed using the Meta-Analysis Accelerator tool[23]. WebPlotDigitizer was used to extract data reported in figures[24].

The online data extraction sheet included study characteristics, baseline data for studies’ populations and outcome measures data. Study characteristics included study name and year, sample size, design, duration of follow-up, and intervention (drug name, IV or intrathecal) and dosage. Population baseline data included age at drug administration (months), age at diagnosis (days), age at end of follow-up (months), gender, weight (Kg), CHOP INTEND score, hammersmith functional motor extended version scale (HFMSE), symptomatic patients at time of treatment, presymptomatic patients at time of treatment, previously treated with other disease-modifying drugs, support treatment (non-invasive ventilation or nasogastric tube). Primary outcome measures data included: overall survival, event-free survival (defined as survival without new onset need of ventilatory support), overall adverse events, serious adverse events, drug-related adverse events, ALT elevation 2x limit of normal or 100 U/L, AST elevation 2x limit of normal or 100 U/L, thrombocytopenia 150 k/ L, vomiting, jaundice, pyrexia, upper respiratory tract infection, pneumonia. Secondary outcome measures included: change from baseline in CHOP-INTEND score and HFMSE; and CHOP-INTEND score ≥40.

Risk of bias assessment:

The risk of bias in the included observational studies was evaluated using the ROBINS-I Version 2 tool[25]. This instrument examines seven specific bias domains: bias due to confounding, bias in the selection of participants into the study, bias in the classification of interventions, bias due to deviations from intended interventions, bias due to missing data, bias in the measurement of outcomes, and bias in the selection of the reported result. An overall judgment of "low," "moderate," "serious," or "critical" risk of bias was determined by the most frequent and highest level of risk identified across these domains. The assessment was conducted independently by 5 authors, and any differences in judgment were resolved through consensus.

Statistical analysis:

To perform our single-arm meta-analysis, we used “meta” package on Rstudio to conduct our random-effect model analysis[26]. Dichotomous outcomes were represented as single proportions and 95% confidence intervals (CI) while continuous outcomes were represented as pooled mean changes from baseline and 95% CI. A statistically significant P-value was considered if it was < 0.05. Heterogeneity was suspected if I2 > 50% by using the Higgins score[27]. A subgroup analysis was done based on different drug doses and prior treatment status.

Literature search

Our systematic search across PubMed, Web of Science, and Scopus initially identified 1230 records. After removing 313 duplicates, a total of 917 studies underwent title and abstract screening. Of these, 94 full texts were reviewed for eligibility, and 21 studies met the inclusion criteria. The PRISMA flow diagram is presented in Figure 1.

Studies and population baseline characteristics

A total of 21 studies were included, encompassing a mix of retrospective cohorts, open-label single-arm trials, and multicenter observational studies, with sample sizes ranging from 3 to 101 participants. Most investigations administered a single intravenous dose of onasemnogene abeparvovec at 1.1 × 10¹⁴ vg/kg over 30–60 minutes, whereas a few early-phase or exploratory studies used intrathecal administration or dose-escalation regimens. Follow-up duration varied between three months and 2 years, depending on study design and patient age at enrollment. Across studies, the mean age at treatment ranged from approximately 3 to 25 months, with baseline weights averaging 6–9 kg. The proportion of male participants typically ranged from 30% to 60%. Baseline CHOP-INTEND scores varied widely—from approximately 16 to 49, reflecting both presymptomatic and symptomatic populations. Several studies included previously treated patients, and supportive care such as noninvasive ventilation or nasogastric feeding was reported in participants in some cohorts. A detailed summary of the included studies and baseline characteristics are provided in Table 1 and Table 2.

Quality assessment

Quality assessment using the ROBINS-I tool indicated that the overall risk of bias among the included studies was predominantly serious. Most studies demonstrated a low risk of bias across domains related to participant selection, classification of interventions, deviations from intended interventions, and missing data. However, several studies were judged to have a moderate to serious risk in the domains of risk of bias due to confounding, measurement of the outcome, and selection of reported results. Consequently, the overall judgment for the majority of studies was serious risk of bias, reflecting inherent limitations in nonrandomized study designs and variability in outcome reporting. The risk of bias graph is shown in Supplementary Figure S1.

Table 1 Summary of the included studies

NA: Not Available .

Study ID	Sample size	Design	Average duration of follow-up	Intervention (IV or intrathecal) and dosage
Chand 2022[28]	101	Retrospective cohort	NA	single dose IV
D’Silva 2022[29]	21	Retrospective cohort	12 months	single dose IV (1.19 10¹⁴ vg/kg) over 60 min.
Day 2021[12]	22	Open-label, single-arm, single-dose, multicentre trial.	18 months	single dose IV 1·1×10¹⁴ vector genomes [vg]/kg for 30–60 min via peripheral vein
Finkel 2023[9]	32	Phase I, open-label, ascending dose study	NA	Intrathecal 3 groups cohort 1: low dose, 6.0 × 10¹³ vg cohort 2: medium dose, 1.2 × 10¹⁴ vg cohort 3: high dose, 2.4 × 10¹⁴ vg
Xiuwei Ma 2024[8]	3	Open-label, single-arm, phase 1 clinical trial	12 months	Intrathecal 2.4 × 10¹⁴ GC101 vector genomes [vg]
Mercuri 2021[13]	33	Multicentre, single-arm, single-dose, open-label phase 3 trial	until end of the study (at age 18 months) if applicable	IV 1·1 ×10¹⁴ vector genomes [vg]/kg for 60 min
Lavie 2025[14]	25	Multicenter retrospective cohort study	12 months	single dose IV 1.1 × 10¹⁴ vg per kg of body weight.
Alves 2024[30]	5	Retrospective	18 months	NA
Bitetti 2023[31]	9	Cohort	3 months	IV 1.1 × 10¹⁴ vg/kg
Desguerre 2024[32]	29	Prospective, multicenter, observational cohort study	24 months	IV (central venous catheter)
Favia 2024[33]	8	Retrospective registry study	6 months	single dose IV 1 x 10¹⁴ vg/kg
Gowda 2024[15]	78	Multicentre, observational cohort study	3 to 22 months	single dose IV 1.1 × 10¹⁴ vg/kg of body weight
Latzer 2023[34]	23	Multicenter observational study	18.0 months	single dose IV 1.1 × 10¹⁴ vg/kg of body weight over one hour
Waldrop 2024[35]	46	Retrospective chart review	7 to 49.5 months.	NA
Mendonça 2024[36]	33	Multi-center retrospective observational study	12 months	IV 1.1 × 10¹⁴ vg/kg of body weight over one hour
Spellman 2025[37]	8	Longitudinal cohort study	3 months	IV 1.1 x 10¹⁴ vg/ kg of body weight
Weiß 2022[38]	51	A multicenter, prospective cohort study	6 months	IV 1·1 × 10¹⁴ vg per kg bodyweight over 1 hour
Mendell 2017[11]	3 (Low dose group)	Cohort study	2 years	Intravenous low dose (6.7×10¹³ vg per kg of body weight).
Mendell 2017[11]	12 (high dose group)	Cohort study	2 years	Intravenous high dose (2.0×10¹⁴ vg per kg of body weight).
Friese 2021[39]	8	Retrospective cohort study	6 months	IV 1.1×10¹⁴ vg/kg
Lee 2022[40]	5	Retrospective cohort study	6 months	IV 1.1 10¹⁴ vg/kg
Ropars 2025[41]	12	Retrospective cohort registry-based study	39.9 months	NA

Table 2 Baseline population characteristics for the included studies.

NA: Not Available .

Study ID	Sample size	Age at drug administration (months) mean (SD)	Age at diagnosis (days) mean (SD)	Age at end of follow-up (months) mean (SD)	Male N (%)	Baseline weight (Kg) mean (SD)	Baseline CHOP INTEND score mean (SD)	Baseline HFMSE mean (SD)	Symptomatic patients at time of treatment N (%)	Presymptomatic patients at time of treatment N (%)	Previously treated with other disease modifying drugs N (%)	Support treatment (noninvasive ventilation) N (%)	Support treatment (nasogastric tube) N (%)
Chand 2022[28]	101	NA	NA	NA	NA	NA	NA	NA	NA	NA	2 (1.98%)	NA	NA
D’Silva 2022[29]	21	11.7 (6.17)	98 (84.7)	24.5 (10.32)	7 (33)	8.15 (2.6)	48.83 (11.77)	25.55 (7.01)	14 (66.6)	0	19 (90.4)	7( 33.3)	7 (33.3)
Day 2021[12]	22	3·7 (1·6)	56·1 (98·6)	NA	10 (45%)	5.8 (0.37)	32·0 (9·7)	NA	22 (100)	0	NA	7 (31.8)	7 (31.8)
Finkel 2023[9]	3 (Low dose group)	17.7 (4.02)	NA	NA	1 (33.3)	9.9 (2.18)	NA	NA	NA	NA	NA	NA	0 (0)
	13 (Medium dose group)	16.35 (4.78)	NA	NA	7 (53.8)	9.5 (0.74)	NA	NA	NA	NA	NA	NA	0 (0)
	4 (High dose group)	16.95 (6.19)	NA	NA	4 (100)	9.1 (0.38)	NA	NA	NA	NA	NA	NA	0 (0)
Xiuwei Ma 2024[8]	3	6.08 (1.73)	NA	NA	0	NA	NA	NA	3 (100%)	0	NA	NA	NA
Mercuri 2021[13]	33	4·1 (1·3)	81·3 (36·4)	18	14 (42%)	5·8 (1)	27·9 (8·3)	NA	33 (100)	0	Na	9 (27 %)	9 (27)
Lavie 2025[14]	25	8.89 (5.76)	5.79 (4.3)	NA	12 (48%)	NA	NA	NA	NA	NA	9 (36)	10 (43%)	5(20)
Alves 2024[30]	5	15.2 (3.31)	NA	31.2 (7.36)	2 (40)	NA	23.8(9.22)	NA	4 (80)	1 (20)	5 (100)	1 (20)	0 (0)
Bitetti 2023[31]	9	25 (18.8)	3 (0.9)	NA	6 (66.67)	NA	41 (10.26)	NA	NA	NA	8 (88.9)	9 (100)	2 (22.2)
Desguerre 2024[32]	29	7.5 (2.64)	NA	NA	10 (34.48)	7.4 (1.27)	27.3 (10.41)	NA	29 (100)	0 (0)	0 (0)	1 (3.45)	0 (0)
Favia 2024[33]	8	3.5 (5.12)	NA	NA	6 (75)	7.5 (2.18)	40.13 (13.36)	NA	NA	NA	6 (75)	1 (12.5)	NA
Gowda 2024[15]	78	NA	NA	NA	NA	NA	NA	NA	NA	NA	54 (54.55)	NA	NA
Latzer 2023[34]	23	8.65 (7.29)	2.23 (1.45)	21.39 (11.04)	14 (60.87)	6.43 (2.13)	30.61 (14.77)	17.5 (6.5)	21 (91.3)	2 (8.7)	9 (39.13)	5 (21.74)	3 (13.04)
Waldrop 2024[35]	46	6.19 (6.95)	NA	39.48 (15.51)	NA	6.08 (2.69)	47.07 (8.13)	NA	19 (41.3)	27 (58.7)	13 (28.26)	9 (19.56)	10 (21.7)
Mendonça 2024[36]	33	16.5 (5.3)	NA	NA	13 (39.4)	9.3 (1.7)	31.2 (13.9)	NA	33 (100)	0(0)	30 (90.9)	29 (87.88)	18 (54.55)
Spellman 2025[37]	8	2.73 (2.1)	NA	14.23 (3.96)	3 (37.5)	NA	NA	NA	0 (0)	8 (100)	0 (0)	NA	NA
Weiß 2022[38]	51	16.27 (10.58)	NA	NA	45 (59)	9.1 (2.29)	NA	NA	70 (92%)	6 (8%)	NA	NA	NA
Mendell 2017[11]	3 (Low dose group)	6.3 (0.75)	33 (46.6)	NA	1 (33)	6.6 (0.63)	16 (12.1)	NA	NA	NA	NA	3 (100)	3 (100)
Mendell 2017[11]	12 (High dose group)	3.4 (2.14)	60 (41.59)	26.07 (3.61)	5 (42)	5.7 (1.47)	28 (11.6)	NA	NA	NA	NA	2 (17)	5 (42)
Friese 2021[39]	8	(10-39)	(1-12)	NA	4 (50)	(7-11.9)	NA	NA	NA	NA	7 (100%)	1 (12.5)	1 (12.5)
Lee 2022[40]	5	10.6 (3.525)	NA	23.3 (2.928)	1 (20)	7.56 (1)	30.8 (10.225)	NA	NA	NA	5 (100%)	2 (40)	1 (20)
Ropars 2025[41]	12	6.1 (3.0)	5.2 (2.9)	NA	7 (58)	6.8 (1.7)	26.1 (6.1)	NA	12 (100)	0	0 (0)	12 (100)	1 (8)

Meta-analysis

Overall and event-free survival

Regarding overall survival, across included studies the pooled survival proportion after onasemnogene treatment was very high. Using a random-effects model, the overall pooled survival was 0.98 (95% CI 0.96–0.99), with low between-study heterogeneity (I² = 9.4%, p = 0.35) as shown in figure 2.

In subgroup analysis by previous treatment status with any disease-modifying agents, studies of previously treated patients showed a pooled survival of 0.98 (95% CI 0.96–1.00) (I² = 24.2%, p = 0.23) as shown in figure 3. In the dose-stratified analysis the standard-dose group had a pooled survival of 0.98 (95% CI 0.96–0.99) (I² = 9.4%, p = 0.35), whereas the high-dose group showed a pooled estimate of 1.00 (95% CI 0.94–1.00) with no observed heterogeneity (I² = 0%, p = 1.00) (Supplementary Figure S2). Overall, heterogeneity was low and there were no significant differences between subgroups, indicating consistently high survival following onasemnogene across study types, prior-treatment status, and dose groups.

The pooled event-free survival after onasemnogene treatment was high but heterogeneous. Interpreting the primary forest plot, the random-effects pooled proportion was 0.78 (95% CI 0.66–0.87). Between-study heterogeneity was substantial (I² = 77.4%, p < 0.0001) (Figure 4). Sensitivity analyses (leave-one-out) did not materially change the pooled estimate or resolve the significant heterogeneity. (Supplementary Figure S3). In subgroup analysis by prior disease-modifying therapy exposure, studies of previously treated patients had a pooled event-free survival of 0.80 (95% CI 0.62–0.93) (I² = 83.8%, p < 0.0001), and no significant subgroup difference existed (Supplementary Figure S4). In the dose-stratified analysis the standard-dose group pooled at 0.78 (95% CI 0.66–0.87) (I² = 77.4%, p < 0.0001), while the high-dose group pooled at 1.00 (95% CI 0.94–1.00) with no observed heterogeneity (I² = 0%, p = 1.00). The test for subgroup differences by dose was significant (p = 0.0005) (Figure 5).

Safety and side effects

Across studies, overall adverse events were common: pooled proportion 0.94 (95% CI 0.75–1.00) with high heterogeneity (I² = 91.8%, p < 0.0001) (Supplementary Figure S5); leave-one-out sensitivity analysis did not resolve the observed heterogeneity. Drug-related adverse events were less frequent: pooled 0.63 (95% CI 0.49–0.75) with low heterogeneity (I² = 21.0%, p = 0.28) (Figure 6).

Elevations in ALT and AST, defined as ≥2× the upper limit of normal or >100 U/L, were analyzed. The pooled proportion of ALT elevation was 0.42 (95% CI 0.24–0.61) with substantial between-study heterogeneity (I² = 81.4%, p < 0.0001) (Supplementary Figure S6); leave-one-out sensitivity analysis did not resolve the observed heterogeneity. The pooled proportion of AST elevation was 0.39 (95% CI 0.21–0.60), also with high heterogeneity (I² = 85.5%, p < 0.0001) and no single-study exclusion resolve the heterogeneity (Supplementary Figure S7). Pneumonia pooled estimate was 0.23 (95% CI 0.09–0.41), and between-study heterogeneity was substantial (I² = 68.3%, p = 0.0076) and heterogeneity was not resolved by sensitivity analysis (Supplementary Figure S8); pyrexia had a pooled proportion of 0.57 (95% CI 0.34–0.78) and heterogeneity was not resolved on sensitivity analysis (Supplementary Figure S9). Serious adverse events pooled at 0.30 (95% CI 0.09–0.57) with high heterogeneity (I² = 81.5%, p = 0.0010) that was not resolved on sensitivity analysis (Supplementary Figure S10). Thrombocytopenia showed pooled estimate of 0.36 (95% CI 0.18–0.58) with very high heterogeneity (I² = 91.1%, p < 0.0001) that was not resolved (Supplementary Figure S11). Upper respiratory tract infection (URTI) pooled at 0.40 (95% CI 0.08–0.78) with very high heterogeneity (I² = 95.1%, p < 0.0001) that was not resolved on sensitivity analysis (Supplementary Figure S12). Finally, vomiting incidence showed a pooled proportion of 0.52 (95% CI 0.17–0.85) with very high heterogeneity (I² = 93.3%, p < 0.0001) that was not resolved on sensitivity analysis (Supplementary Figure S13).

CHOP-INTEND and HFMS scales

The proportion of patients achieving a CHOP-INTEND score ≥40 after onasemnogene treatment was high where the pooled estimate was 0.91 (95% CI 0.81–0.98), with moderate heterogeneity (I² = 61.5%, p = 0.0077) (Figure 7).

In leave-one-out sensitivity analysis, the pooled estimate remained favorable, and the heterogeneity was resolved (I² = 49.0%) by omitting Mercuri et al. 2021 (Supplementary Figure S14). Mean change from baseline in CHOP-INTEND scale showed a pooled effect estimate of 15.77 (95% CI 12.07–19.47), although between-study heterogeneity was high (I² = 91.8%, p < 0.0001) (Figure 8). Leave-one-out sensitivity analysis did not resolve the substantial heterogeneity. Subgrouping based on the status of previous treatment showed a statistically significant subgroup difference favoring the studies including patients who did not receive any previous treatments (p = 0.0817) (Supplementary Figure S15). Regarding HFMSE, mean change from baseline showed a pooled effect estimate of 10.98 (95% CI 8.42–13.53). Between-study heterogeneity was negligible (I² = 0.0%, p = 0.9366) (Supplementary Figure S16).

The approval of onasemnogene abeparvovec (OA) in May 2019 [42]was based on a limited number of clinical trials with restrictive enrollment criteria, focusing on patients younger than 8 months, weighing less than 8 kg, and who were treatment-naive. Consequently, controversy arose among neuromuscular experts regarding the efficacy and safety of OA in older, heavier patients, those previously treated with nusinersen, and individuals with bulbar or respiratory impairment[43].

To address this controversy, our systematic review and meta-analysis synthesized data from retrospective cohorts, open-label single-arm trials, and multicenter observational studies, encompassing 565 participants with SMA1. We found that OA was associated with significantly improved survival and motor function compared to natural history data[44, 45]. Importantly, OA demonstrated a tolerable safety profile across a diverse range of baseline characteristics.

The pooled overall survival was very high across included studies 98% (95% CI 0.96–0.99, I² = 9.4%, n = 565). Our findings are consistent with prior meta-analyses by Fernandes et al.[19] 97.56% (95%CI: 92.55–99.86, I2 = 0%, n = 67), However, our study included a larger sample size and employed a random-effects model, providing more robust estimates.

We conducted subgroup analyses to explore specific controversies. A subgroup analysis based on previous treatment with other disease-modifying agents showed a non-significant trend favoring those who were pre-treated (p = 0.098), which may reflect differences in the included study populations. The random-effects pooled proportion for event-free survival was 78% (95% CI: 0.66–0.87), which is lower than the 96.5% reported by Fernandes et al[19]. This discrepancy is likely explained by our inclusion of older, heavier patients with more advanced disease. Furthermore, a subgroup analysis based on OA dosage suggested that a higher dose may be associated with a promising 100% event-free survival, though this is based on a small sample (n = 16). The clinical significance of these findings is stark when compared to natural history cohort studies. The clinical significance of these findings is stark when compared to natural history cohort studies. In these cohorts, patients with two copies of SMN2 typically reach the overall median survival time to permanent ventilatory support or death by 8 to 10.5 months of age[44, 45].

Motor function assessments reinforced OA's benefit. The estimated proportion of patients achieved CHOP-INTEND score ≥ 40 was 91%, approximate to the findings of Fernandes et al., 87.28% and published systemic reviews[19, 46, 47]. Moreover, the mean change from baseline in CHOP-INTEND scale showed a pooled effect estimate of 15.77 (95% CI 12.07–19.47, n = 177)), with statistical difference across studies favoring treatment naive patients 20.96(95% CI 19.14–22.87, n = 39). Regarding HFMSE, mean change from baseline showed a pooled effect estimate of 10.98 (95% CI 8.42–13.53, n = 11). Conversely, none of participants of natural history SMA1 accept one motor skills as the motor function affected progressively over disease course[19, 45].

The analysis clearly observed the early OA administration produce better result of motor skill achievements scores. The pretreated individuals were already older than and with advanced disease compared with treatment-naive patient[14, 15, 31, 34–36]. These results are supported by studies that administered the gene therapy prior to 6 weeks of age and before the onset of symptoms[12, 13, 33]. Although OA demonstrates motor improvement in symptomatic and advanced-stage SMA1 individuals—including those who were older and had more advanced disease at baseline compared to treatment-naïve cohorts—these gains are notably delayed and less pronounced than those achieved with younger administration[9, 14, 30–32].

In the natural history cohort, all (100%) patients older than 12 months required either nutritional support or combined nutritional and ventilatory support[19, 45]. In contrast, the early administration of onasemnogene abeparvovec showed promising results that between 82–100% of included patients did not use ventilator support at age 12 and 18 months[9, 12, 13]. These observations draw attention to the previously reported issue of early diagnosis, ideal time of administering OA therapy, and newborn screening program importance for early intervention.[48]

The comprehensive safety and adverse events profile for onasemnogene abeparvovec incorporates post-commercialization data and specifically addresses five key adverse events of special interest: hepatotoxicity, thrombocytopenia, cardiac events, thrombotic microangiopathy, and ganglionopathy)[49]. The evaluation of the safety of onasemnogene abeparvovec for a wider range of patients observed in this review showed pooled proportion of 94% of any adverse event reported. Drug-related adverse events were more frequent with 63% proportion compared with Fernandes et al[19], 52.64%.

Hepatotoxicity typically presents as non-cholestatic (i.e., as increases in serum aminotransferase concentrations) and most often occurs at 1 week and 1 month after treatment[50]. The acute increase of liver enzyme concentration within first week in our review were estimated with pooled proportion 42%, 39% for ALT and AST respectively. A subacute elevation of transaminases between 2–5 weeks after gene replacement therapy were reported in prolonged observational studies[38, 39]. These results were supported with previous reported data, which made it necessary to use prednisolone combined with onasemnogene abeparvovec to prevent possibility of acute liver failure[51].

The most frequently reported drug-related adverse events are included in the product labelling of onasemnogene abeparvovec, including thrombocytopenia (36%), vomiting (52%), and pyrexia (57%). The clinical manifestations of sensory ganglionopathy were suggested but not confirmed in the studies where OA was administered intrathecally[9] and were not reported for intravenous administration.[13, 28]

The studies included in this meta-analysis have significant methodological limitations, including a high risk of bias, small sample sizes, and a lack of control groups, which precludes randomization and complicates statistical summarization. We observed substantial heterogeneity (I² >75%) for several outcomes, including event-free survival and adverse event rates, indicating that the pooled estimates should be interpreted with caution. The rarity of SMA1 makes large, randomized trials challenging. While previous reviews have used indirect comparison techniques, such as Matching-Adjusted Indirect Comparison (MAIC) and Simulated Treatment Comparison (STC), they proved inconclusive with regard to some outcomes or identified imprecise results due to large confidence intervals.[52] A key strength of our study is its larger sample size and the ability to perform comparative subgroup analyses.

This review reinforces that onasemnogene abeparvovec significantly improves survival and motor function in SMA1 patients, with a manageable safety profile. Crucially, it demonstrates efficacy and tolerability in a broader patient population than originally approved, including older individuals and those with more advanced disease. However, the magnitude of benefit is greatest with early intervention. These findings highlight the paramount importance of newborn screening and prompt treatment while supporting the use of OA in a wider range of patients, including those who have received prior nusinersen therapy.

Acknowledgments: None.

Authors' contributions: The authors meet the criteria for authorship as recommended by the International Committee of Medical Journal Editors (ICMJE).
S.M.A, G.H and H.E.M contributed equally to this work and designated as co-first authors. S.M.A, G.H and H.E.M contributed to the study’s conception and design. M.N, M.E.H, H.Y, M.K.D and S.Y performed studies screening. M.N, M.E.H, H.Y, S.Y, and M.K.D performed data extraction. S.M.A, G.H, H.E.M, and H.Y wrote the initial draft of the manuscript. S.M.A and G.H revised and prepared the manuscript for submission. All authors provided feedback on earlier drafts of the manuscript. The final manuscript was read and approved by all authors.

Statements and Declarations:

Ethical considerations: Not applicable

Consent to participate: Not applicable

Consent for publication: Not applicable

Declaration of conflicting interest: The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding statement: None

Data availability: All data generated or analyzed during this study are included in this published article [and its supplementary information file].

Prior TW, Finanger E (1993) Spinal Muscular Atrophy. In: Pagon RA, Adam MP, Ardinger HH, et al (eds) GeneReviews(®). University of Washington, Seattle, Seattle (WA)
Farrar MA, Kiernan MC (2015) The genetics of spinal muscular atrophy: progress and challenges. Neurotherapeutics 12:290–302. https://doi.org/10.1007/s13311-014-0314-x
Verhaart IEC, Robertson A, Wilson IJ, et al (2017) Prevalence, incidence and carrier frequency of 5q-linked spinal muscular atrophy - a literature review. Orphanet J Rare Dis 12:124. https://doi.org/10.1186/s13023-017-0671-8
Ogbonmide T, Rathore R, Rangrej SB, et al (2023) Gene therapy for spinal muscular atrophy (SMA): A review of current challenges and safety considerations for onasemnogene abeparvovec (zolgensma). Cureus 15:e36197. https://doi.org/10.7759/cureus.36197
Schwartz M, Likhite S, Meyer K (2021) Onasemnogene abeparvovec-xioi: a gene replacement strategy for the treatment of infants diagnosed with spinal muscular atrophy. Drugs Today 57:387–399. https://doi.org/10.1358/dot.2021.57.6.3264117
Foust KD, Nurre E, Montgomery CL, et al (2009) Intravascular AAV9 preferentially targets neonatal neurons and adult astrocytes. Nat Biotechnol 27:59–65. https://doi.org/10.1038/nbt.1515
Blair HA (2022) Onasemnogene abeparvovec: A review in spinal muscular atrophy. CNS Drugs 36:995–1005. https://doi.org/10.1007/s40263-022-00941-1
Ma X, Zhuang L, Ma W, et al (2024) Treatment of SMA type 1 infants using a single-dose AAV9-mediated gene therapy via intrathecal injection of GC101: An open-label, single-arm study. Chin Med J 137:1976–1978. https://doi.org/10.1097/CM9.0000000000003210
Finkel RS, Darras BT, Mendell JR, et al (2023) Intrathecal Onasemnogene Abeparvovec for Sitting, Nonambulatory Patients with Spinal Muscular Atrophy: Phase I Ascending-Dose Study (STRONG). J Neuromuscul Dis 10:389–404. https://doi.org/10.3233/JND-221560
Novartis intrathecal onasemnogene abeparvovec Phase III study meets primary endpoint in children and young adults with SMA | Novartis. https://www.novartis.com/news/media-releases/novartis-intrathecal-onasemnogene-abeparvovec-phase-iii-study-meets-primary-endpoint-children-and-young-adults-sma. Accessed 31 Oct 2025
Mendell JR, Al-Zaidy S, Shell R, et al (2017) Single-Dose Gene-Replacement Therapy for Spinal Muscular Atrophy. N Engl J Med 377:1713–1722. https://doi.org/10.1056/NEJMoa1706198
Day JW, Finkel RS, Chiriboga CA, et al (2021) Onasemnogene abeparvovec gene therapy for symptomatic infantile-onset spinal muscular atrophy in patients with two copies of SMN2 (STR1VE): an open-label, single-arm, multicentre, phase 3 trial. Lancet Neurol 20:284–293. https://doi.org/10.1016/S1474-4422(21)00001-6
Mercuri E, Muntoni F, Baranello G, et al (2021) Onasemnogene abeparvovec gene therapy for symptomatic infantile-onset spinal muscular atrophy type 1 (STR1VE-EU): an open-label, single-arm, multicentre, phase 3 trial. Lancet Neurol 20:832–841. https://doi.org/10.1016/S1474-4422(21)00251-9
Lavie M, Rochman M, Armoni Domany K, et al (2024) Respiratory outcomes of onasemnogene abeparvovec treatment for spinal muscular atrophy: national real-world cohort study. Eur J Pediatr 184:58. https://doi.org/10.1007/s00431-024-05886-9
Gowda V, Atherton M, Murugan A, et al (2024) Efficacy and safety of onasemnogene abeparvovec in children with spinal muscular atrophy type 1: real-world evidence from 6 infusion centres in the United Kingdom. Lancet Reg Health Eur 37:100817. https://doi.org/10.1016/j.lanepe.2023.100817
Daghriri S, Alorfi YA, Alayed RS, et al (2025) The efficacy and safety of Nusinersen for spinal muscular atrophy types 1, 2, 3: a systematic review of the current evidence. Egypt J Neurol Psychiatry Neurosurg 61:30. https://doi.org/10.1186/s41983-025-00959-4
Kokaliaris C, Evans R, Hawkins N, et al (2024) Long-Term Comparative Efficacy and Safety of Risdiplam and Nusinersen in Children with Type 1 Spinal Muscular Atrophy. Adv Ther 41:2414–2434. https://doi.org/10.1007/s12325-024-02845-6
Masson R, Mazurkiewicz-Bełdzińska M, Rose K, et al (2022) Safety and efficacy of risdiplam in patients with type 1 spinal muscular atrophy (FIREFISH part 2): secondary analyses from an open-label trial. Lancet Neurol 21:1110–1119. https://doi.org/10.1016/S1474-4422(22)00339-8
Fernandes BD, Krug BC, Rodrigues FD, et al (2024) Efficacy and safety of onasemnogene abeparvovec for the treatment of patients with spinal muscular atrophy type 1: A systematic review with meta-analysis. PLoS ONE 19:e0302860. https://doi.org/10.1371/journal.pone.0302860
Page MJ, McKenzie JE, Bossuyt PM, et al (2021) The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ n71. https://doi.org/10.1136/bmj.n71
Bramer WM, Giustini D, de Jonge GB, et al (2016) De-duplication of database search results for systematic reviews in EndNote. J Med Libr Assoc 104:240–243. https://doi.org/10.3163/1536-5050.104.3.014
Ouzzani M, Hammady H, Fedorowicz Z, Elmagarmid A (2016) Rayyan-a web and mobile app for systematic reviews. Syst Rev 5:210. https://doi.org/10.1186/s13643-016-0384-4
Abbas A, Hefnawy MT, Negida A (2024) Meta-analysis accelerator: a comprehensive tool for statistical data conversion in systematic reviews with meta-analysis. BMC Med Res Methodol 24:243. https://doi.org/10.1186/s12874-024-02356-6
automeris.io: Computer vision assisted data extraction from charts using WebPlotDigitizer. https://automeris.io/. Accessed 31 Oct 2025
Risk of bias tools - ROBINS-I V2 tool. https://www.riskofbias.info/welcome/robins-i-v2. Accessed 31 Oct 2025
Schwarzer G, Carpenter JR, Rücker G (2015) Meta-Analysis with R. Springer International Publishing, Cham
Higgins JPT, Thompson SG, Deeks JJ, Altman DG (2003) Measuring inconsistency in meta-analyses. BMJ 327:557–560. https://doi.org/10.1136/bmj.327.7414.557
Chand DH, Mitchell S, Sun R, et al (2022) Safety of onasemnogene abeparvovec for patients with spinal muscular atrophy 8.5 kg or heavier in a global managed access program. Pediatr Neurol 132:27–32. https://doi.org/10.1016/j.pediatrneurol.2022.05.001
D’Silva AM, Holland S, Kariyawasam D, et al (2022) Onasemnogene abeparvovec in spinal muscular atrophy: an Australian experience of safety and efficacy. Ann Clin Transl Neurol 9:339–350. https://doi.org/10.1002/acn3.51519
Alves BKAM de F, Araujo AP de QC, Santos FND, Ribeiro MG (2024) Type-1 spinal muscular atrophy cohort before and after disease-modifying therapies. Arq Neuropsiquiatr 82:1–8. https://doi.org/10.1055/s-0044-1791757
Bitetti I, Lanzara V, Margiotta G, Varone A (2023) Onasemnogene abeparvovec gene replacement therapy for the treatment of spinal muscular atrophy: a real-world observational study. Gene Ther 30:592–597. https://doi.org/10.1038/s41434-022-00341-6
Desguerre I, Barrois R, Audic F, et al (2024) Real-world multidisciplinary outcomes of onasemnogene abeparvovec monotherapy in patients with spinal muscular atrophy type 1: experience of the French cohort in the first three years of treatment. Orphanet J Rare Dis 19:344. https://doi.org/10.1186/s13023-024-03326-3
Favia M, Tarantino D, Cerbo LD, et al (2024) Onasemnogene Abeparvovec: Post-infusion Efficacy and Safety in Patients With Spinal Muscular Atrophy (SMA)-A Fondazione Policlinico Gemelli IRCCS Experience. Hosp Pharm 59:39–46. https://doi.org/10.1177/00185787231182562
Tokatly Latzer I, Sagi L, Lavi R, et al (2023) Real-Life Outcome After Gene Replacement Therapy for Spinal Muscular Atrophy: A Multicenter Experience. Pediatr Neurol 144:60–68. https://doi.org/10.1016/j.pediatrneurol.2023.04.007
Waldrop MA, Chagat S, Storey M, et al (2024) Continued safety and long-term effectiveness of onasemnogene abeparvovec in Ohio. Neuromuscul Disord 34:41–48. https://doi.org/10.1016/j.nmd.2023.11.010
Mendonça RH, Ortega AB, Matsui C, et al (2024) Gene replacement therapy for spinal muscular atrophy: safety and preliminary efficacy in a Brazilian cohort. Gene Ther 31:391–399. https://doi.org/10.1038/s41434-024-00456-y
Spellman RG, Ha LL, Da Silva Duarte Lepez S, et al (2025) Early life safety profiling of gene therapy for spinal muscular atrophy. Gene Ther. https://doi.org/10.1038/s41434-025-00529-6
Weiß C, Ziegler A, Becker L-L, et al (2022) Gene replacement therapy with onasemnogene abeparvovec in children with spinal muscular atrophy aged 24 months or younger and bodyweight up to 15 kg: an observational cohort study. Lancet Child Adolesc Health 6:17–27. https://doi.org/10.1016/S2352-4642(21)00287-X
Friese J, Geitmann S, Holzwarth D, et al (2021) Safety Monitoring of Gene Therapy for Spinal Muscular Atrophy with Onasemnogene Abeparvovec -A Single Centre Experience. J Neuromuscul Dis 8:209–216. https://doi.org/10.3233/JND-200593
Lee S, Lee YJ, Kong J, et al (2022) Short-term clinical outcomes of onasemnogene abeparvovec treatment for spinal muscular atrophy. Brain Dev 44:287–293. https://doi.org/10.1016/j.braindev.2021.12.006
Ropars J, Cances C, Garcia-Uzquiano R, et al (2025) Comparative clinical outcomes of nusinersen and gene therapy in spinal muscular atrophy type 1. JAMA Netw Open 8:e2536348. https://doi.org/10.1001/jamanetworkopen.2025.36348
Hoy SM (2019) Onasemnogene abeparvovec: first global approval. Drugs 79:1255–1262. https://doi.org/10.1007/s40265-019-01162-5
René CA, Parks RJ (2023) Expanding the availability of onasemnogene abeparvovec to older patients: the evolving treatment landscape for spinal muscular atrophy. Pharmaceutics 15:. https://doi.org/10.3390/pharmaceutics15061764
Kolb SJ, Coffey CS, Yankey JW, et al (2017) Natural history of infantile-onset spinal muscular atrophy. Ann Neurol 82:883–891. https://doi.org/10.1002/ana.25101
Finkel RS, McDermott MP, Kaufmann P, et al (2014) Observational study of spinal muscular atrophy type I and implications for clinical trials. Neurology 83:810–817. https://doi.org/10.1212/WNL.0000000000000741
Yang D, Ruan Y, Chen Y (2023) Safety and efficacy of gene therapy with onasemnogene abeparvovec in the treatment of spinal muscular atrophy: A systematic review and meta-analysis. J Paediatr Child Health 59:431–438. https://doi.org/10.1111/jpc.16340
Pascual-Morena C, Cavero-Redondo I, Lucerón-Lucas-Torres M, et al (2023) Onasemnogene Abeparvovec in Type 1 Spinal Muscular Atrophy: A Systematic Review and Meta-Analysis. Hum Gene Ther 34:129–138. https://doi.org/10.1089/hum.2022.161
Serra-Juhe C, Tizzano EF (2019) Perspectives in genetic counseling for spinal muscular atrophy in the new therapeutic era: early pre-symptomatic intervention and test in minors. Eur J Hum Genet 27:1774–1782. https://doi.org/10.1038/s41431-019-0415-4
Day JW, Mendell JR, Mercuri E, et al (2021) Clinical trial and postmarketing safety of onasemnogene abeparvovec therapy. Drug Saf 44:1109–1119. https://doi.org/10.1007/s40264-021-01107-6
Chand D, Mohr F, McMillan H, et al (2021) Hepatotoxicity following administration of onasemnogene abeparvovec (AVXS-101) for the treatment of spinal muscular atrophy. J Hepatol 74:560–566. https://doi.org/10.1016/j.jhep.2020.11.001
ZOLGENSMA (onasemnogene abeparvovec) and Fatal Cases of Acute Liver Failure - Canada.ca. https://recalls-rappels.canada.ca/en/alert-recall/zolgensma-onasemnogene-abeparvovec-and-fatal-cases-acute-liver-failure. Accessed 4 Nov 2025
Bischof M, Lorenzi M, Lee J, et al (2021) Matching-adjusted indirect treatment comparison of onasemnogene abeparvovec and nusinersen for the treatment of symptomatic patients with spinal muscular atrophy type 1. Curr Med Res Opin 37:1719–1730. https://doi.org/10.1080/03007995.2021.1947216

The authors declare no competing interests.

Supplementarymaterial.docx

Therapeutic Efficacy and Safety of onasemnogene abeparvovec Gene Therapy for Spinal Muscular Atrophy Type 1: Updated Systematic Review with Meta-Analysis

Status:

Version 1

Abstract

Objectives:

Background:

Methods:

Results:

Conclusion:

Figures

Introduction

Methods

Results

Discussion

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1